The study used ChatGPT to summarize 10 peer-reviewed abstracts each from 14 journals published in 2022. Physicians read the articles and compared them to the summaries produced by ChatGPT and rated the summaries for quality, accuracy and level of bias.
The generated summaries were 70% shorter than the abstracts and were found to be high in quality and accuracy and low in bias, according to a July 17 university news release. There were four instances in which ChatGPT produced text that was not fact based and 20 instances of minor inaccuracies that did not change the overall meaning.
“This study shows us that these tools already have some ability to help us review the literature a little bit faster, as well as figure out where we need to focus our attention,” lead study author Daniel Parente, MD, PhD, assistant professor of family medicine and community health at KU Medical Center, said in the release. “And it seems very likely that future versions of these technologies that are smarter and more capable will only enhance that.”
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
